Erik-JanQixiang of the @odissei_soda team presenting on how to collect data / transform unstructured text into structured ratings using LLMs.
Next up: how to correct for various (measurement) errors in these ratings!
@odissei_soda using the great new #rstats package #ellmer and #python package #langchain
I'll post the url for the materials (slides and code notebooks) soon
@odissei_soda as promised, here are the materials for our workshop!
https://sodascience.github.io/workshop_llm_data_collection/
#OpenSource #OpenScience and everything